RRA (Retrospective Reasoning Analysis)
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Contemporary AI inference systems produce two distinct outputs during operation: the visible response directed at the user, and the latent reasoning trace that precedes and generates that response... We propose Retrospective Reasoning Analysis (RRA) as a framework with two applications: a near-term Directed Reasoning application operating at inference time, and a longer-term Internal Cognition application in which systematic reasoning traces become training signal for building models with self-reflection as a native architectural capacity.
Fields of Study & Subjects
Explainable Artificial Intelligence (XAI)
AI-based Problem Solving and Planning
Logic
Reasoning
and Knowledge
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 26 March 2026
- Language
- EN